Part of the Maths & Statistics suite · 235 calculators

Median Absolute Deviation Calculator

The most robust spread measure there is — the median of the distances from the median, which half your data can be corrupted before it moves.

Take the median, find each value's absolute distance from it, then take the median of those distances.

Comma-separated
Results update as you type
Results
Median absolute deviation
2
Median
Robust σ estimate (1.4826 × MAD)
Sample standard deviation
How much bigger the SD is
Values beyond 3 robust σ
Count
Reviewed September 2026. Pure mathematics: the result does not depend on where you are. Terminology follows the national curriculum (maths, brackets, decimal point).
No account required · Google Analytics off unless allowedCalculator arithmetic runs in your browserResults update as you type
All calculations run 100% in your browser. The calculator code does not submit your figures to GlobalCalc to obtain a result.
About median absolute deviation

How the median absolute deviation calculator works

Take the median, find each value's absolute distance from it, then take the median of those distances. Because both steps are medians, the result has a breakdown point of 50%: it takes more than half the data being wrong to break it.

Multiplying by 1.4826 scales it to estimate σ for normal data, which is how it is used as a drop-in robust replacement for the standard deviation.

Formula: MAD = median(|xᵢ − median(x)|); σ̂ = 1.4826 × MAD

Worked examples

InputsMedian absolute deviationNote
Seven ordinary values and one of 10002MAD = 2, SD = 352 — the outlier is caught
Clean data2MAD = 2, robust σ ≈ 2.97
Identical values0MAD = 0

Frequently asked questions

What is the median absolute deviation?

The median of the absolute distances from the median — a spread measure that ignores extremes completely.

Why 1.4826?

It is 1 ÷ Φ⁻¹(0.75), the factor that makes the MAD estimate σ for normally distributed data.

What is a breakdown point?

The fraction of the data that can be arbitrarily corrupted before the statistic becomes meaningless. The mean's is 0%; the MAD's is 50% — the highest possible.

When should I use it?

Any time outliers are plausible and you cannot inspect every point — sensor data, log timings, scraped measurements.

Is it the same as mean absolute deviation?

No, and the shared acronym causes real confusion. That one averages distances from the mean; this one takes the median of distances from the median.

Where these figures come from

Last checked: September 2026. Formulas are fixed by mathematics and do not change with tax years or regulations.